[arthur pappas]

I really had metering difficulty the other night while snapping shots at a local game. My shots were either washed out & too bright, or simply too dark. I tried different metering settings but my surroundings constantly got the best of me

This shot closely represents real brightness inside the arena


As you can see the ice surface is way overexposed


Trying to cut back on the ice surface overexposure starts to really darken the arena


Finally getting the ice surface to look close to natural but this leaves the arena almost pitch black


What techniques can I use next time I'm at a game to avoid this situation? My friend was using a rediculously old and cheap Kodak P&S camera (3 MP) and it took great shots in full auto mode. Using the Auto or P modes on the K100 was an even bigger disaster...

[flowerpot]

its not the metering problem.. its simply the dynamic range...

[soccerjoe5]

that's a very high contrast scene. that's where we're limited in dynamic range as opposed to film.

the last photo on my screen (calibrated) looks very dark and underexposed. try a half stop higher. the ice/rink area looks too dark for me.

shoot full manual in situations like these to really control the exposure

[piotro]

Well, as it was said above, contrasty scene will give you that sort of problem on DSLR.
You can tweak the last one in software and get much better results (with a bit of overexposing the most bright places, here some example done in quick on a non-calibrated monitor (with a bit stronger jpeg compression, so will lose some detail) .

Attached Images

View Picture EXIF

File Date

 

[spyglass]

Hey great to see another Wackian on the site. It is not like shooting on the old barn (Coliseum) back in the Chief days. Prospera Center can be a bit harsh and unfriendly for shooting in. The ice region meters very high, while the crowd comes in very low. I find that I have to set and shoot one or the other. The only way you can get both is to set the exposure for the ice, and flash fill the stands. I shoot some lacrosse games in Prospera (concrete dry floor being a tad easier than the stark white of an ice sheet). I fell into the habit of framing and shooting just the playing area. It meant that I had to move a great deal (easier at a lacrosse game than a Bruin's game). The other thing I found was the colour rendering of the HID (high intensity discharge lighting / metal halide lamps). If the camera was left on auto white balance the results were average at best (maybe due to the vast amount of grey concrete). I had better results selecting the WB manually. A lens hood helped for some of the stray glare. There just seems to be so many reflective surfaces around the arena (signs, ducts, bald heads - ha ha, etc). As Flowerpot suggested, the dynamic range of the location is beyond what a camera can handle. I am surprised that the point and shoot did ok. (was its results dynamically flat)

The Bruins did well in their last 3 games. (3 ticks in the "w" column). Stand look a little thin.

[Lowell Goudge]

I agree this is a dynamic range issue, and also, a cheap old P&S in full auto mode will also shoot off the flash, providing better foreground exposure, offsetting the influence of the ice rink,

Fill flash can do a lot in situations like this. but you need to always keep in mind that the ice rink is the worst situation (other than being on a snowy field or white sand beach to take shots)

[alohadave]

The other thing I found was the colour rendering of the HID (high intensity discharge lighting / metal halide lamps). If the camera was left on auto white balance the results were average at best (maybe due to the vast amount of grey concrete). I had better results selecting the WB manually.

Even setting WB manually may not solve all the color problems. You can take two pictures right after each other, and one will be white and the next orange. From what I've heard about these lights, it has to do with the frequency of the electrical current.

I had good success taking WB off the ice the last time I shot a game, but times before that were a mix of white and orange shots.

[egordon99]

Ditto.....The last two shots tell the entire story....

its not the metering problem.. its simply the dynamic range...

Anytime you find yourself saying "If I make it so Subject A is properly exposed, subject B is too dark (or blown-out)" you have exceeded the dynamic range of the camera.

Just get a longer lens so you can just "focus" on the players. Who wants to see the crowd anyway?

[alohadave]

Just get a longer lens so you can just "focus" on the players. Who wants to see the crowd anyway?

They are fun to watch.

[Marc Sabatella]

My friend was using a rediculously old and cheap Kodak P&S camera (3 MP) and it took great shots in full auto mode.

If it produced a picture in which you could see plenty of detail in both the crowd and the ice, then by most definitions, the shot was *terrible*, because it doesn't reflect reality. The scene itself, as others have said, exceeds the dynamic range of what can be represented realistically on a two-dimensional surface.

So you're friend's camera may have applied *very* heavy-handed processing in camera to take a picture like you "underexposed one" and artificially bring up the levels in the shadows, or it might have used flash to illuminate the crowd (which of course you could have done as well. Or it might just represent all scenes as containing much less contrast than they really do - which would actually suck in most situations.

[audiobomber]

Even setting WB manually may not solve all the color problems. You can take two pictures right after each other, and one will be white and the next orange. From what I've heard about these lights, it has to do with the frequency of the electrical current.

The frequency of electrical current in Canada is always 60 Hz. The differing white balance has to do with the wavelength of the lighting. Clearly they are using two different types of lamps in the arena. You could shoot in raw under artificial light, and adjust colour (lighting temperature) in post-processing.

[Lowell Goudge]

Even setting WB manually may not solve all the color problems. You can take two pictures right after each other, and one will be white and the next orange. From what I've heard about these lights, it has to do with the frequency of the electrical current.

I had good success taking WB off the ice the last time I shot a game, but times before that were a mix of white and orange shots.

The problem is NOT the frequency of the electrical current, as lights do not change color with frequency. Color is determined by physical properties, temperature (and as a result voltage and current) envelope material and phosphorous coating in the case of incandescent lights, or the gas and phosphorous coating on arc discharge lamps

If you set WB manually, it will be consistent as long as the lighting is, but note that in many buildings they mix the types of lamps to give a more natural color. It is not uncommon to see both "warm white" and "Cool White" flourescent tubes in a school gym, or different color street lamps between high and low pressure sopdium, or mercury discharge lamps. The same holds true for arenas, although I would suspect that for a professional venu they use what the TV crews tell them to for the best TV image regardless of how photo's look,

[soccerjoe5]

if it's artificial lighting, i almost always shoot with custom WB. it's really easy to do with the K20D

[Quicksand]

Maybe I'm splitting hairs here, but it is entirely possible for white balance to change because of the frequency of the lighting power.

Let me give an example.

If an area is lit with a combination of incandescent and gas discharge lighting (like fluorescent, arc, or HID) using 60 Hz current, then you'll see two things:

The incandescent lamps will NOT flicker (much) and will provide pretty much continuous, warm, white balance. The light there comes from the heated filament of the bulb, which does not fluctuate much as the power signal cycles.

On the other hand, gas discharge lighting will flicker at usually twice the frequency of the power, i.e. 120 Hz in North America, because the lamp lights at both positive-going and negative-going extremes of the waveform.

So if you take shots at high shutter speeds (say, faster than 1/250), you might get some shots that occur at the peak of the gas discharge lighting curve, and some that occur at the valleys when the incandescent lighting prevails (in between flickers, if you will).

This will mess with both metering and white balance. But at slower shutter speeds, you shouldn't notice it at all.

[Lowell Goudge]

Maybe I'm splitting hairs here, but it is entirely possible for white balance to change because of the frequency of the lighting power.

Let me give an example.

If an area is lit with a combination of incandescent and gas discharge lighting (like fluorescent, arc, or HID) using 60 Hz current, then you'll see two things:

The incandescent lamps will NOT flicker (much) and will provide pretty much continuous, warm, white balance. The light there comes from the heated filament of the bulb, which does not fluctuate much as the power signal cycles.

On the other hand, gas discharge lighting will flicker at usually twice the frequency of the power, i.e. 120 Hz in North America, because the lamp lights at both positive-going and negative-going extremes of the waveform.

So if you take shots at high shutter speeds (say, faster than 1/250), you might get some shots that occur at the peak of the gas discharge lighting curve, and some that occur at the valleys when the incandescent lighting prevails (in between flickers, if you will).

This will mess with both metering and white balance. But at slower shutter speeds, you shouldn't notice it at all.

this is not a function of frequency of power but more a question of shutter speed relitive to power frequency.

If what you state is truely the cause of inconssitent WB, then you would see banding on your shot as the lights strobe with the shutter opening, (you are implying the light level is pulsating at 120 Hz) I see no evidence of banding in the photos presented to indicate that there is any strobe effect, therefore either it is not taking place or the shutter is too slow to register it (in fact it is the latter as all shots are 1/90th or below)

If you really believe the flicker of lights is an issue, photograph it. It would be a great discussion point but I don't think it is any reason behind the hockey photos.

I went back and looked at shots taken at the ACC at between 1/125 and 1/750th of a second, and there is no variation of exposure across the frame, or banding to suggest the light level is changing within the cycle of commercial power, there is also no change in WB from shot to shot and they were all taken set to daylight.

The same is true for shots taken in school gyms under flourscent lighting.